НАУКА

Description

According to Federal Highway Administration, up to 70% of weather-related accidents happen on wet pavement or during rainfall. Water-pavement interaction is one of the most challenging road safety problems. Rainfall is drained from the pavement surface in order to prevent aquaplaning and provide skidding resistance. Efficiency of surface drainage depends on numerous factors: pavement surface geometry, rainfall intensity and duration, pavement hydraulic properties etc. Environmental issues add an additional complexity. Pollution is moved with water and the process is known as build-up and wash-off phenomena, depending on additional factors such as: traffic load, antecedent dry period, pavement state in terms of clogging etc. Those important issues related to water–pavement interactions are not well-addressed in contemporary science. Current practice relies on experience, standards and oversimplified formulas. In order to tackle all aspects and complexity of this problem, robust next-gen Road and Environmental Safety Toolkit (REST) will be developed. REST will consist of two components: the first one is a complex numerical finite volume model for shallow water flow with infiltration, and the second one is much faster, surrogate AI model that will be trained with data generated from complex numerical model. Project goal is to use REST as a tool to thoroughly investigate and initiate improvements of road and environmental safety measures meaning: revision of current design standards and methods related to road geometry and water film depth calculation, recommendations for using porous asphalt on critical road sections, identification of critical sections on existing road network, producing and publicly presenting implementation plan for improvement of safety of national road network. Long term goal of the project is application of REST for safety forecast in critical weather conditions, considering climate changes, and integration in the driver’s information system.